Three full-length practice tests for the Oracle Agentic AI Foundations Associate (1Z0-1157-26) certification exam - 40 questions per test, 60 minutes, 65% to pass, unlimited retakes.
Every question is grounded in Oracle's official exam objectives and product documentation - written fresh for this course, not recycled from public dumps.
What you'll be tested on
- Core AI agent concepts: autonomy, the LLM-tools-orchestration loop, Chain-of-Thought and ReAct reasoning, and layered guardrails
- Building agents with LangChain and the LangChain Expression Language (LCEL)
- Model Context Protocol (MCP): hosts, clients, servers, and tool discovery
- The OpenAI Agents SDK and Responses API: multi-step logic, multi-agent handoffs, and tracing
- Building and deploying agents on the OCI Enterprise AI platform
- Agentic AI for Oracle AI Database: Select AI, Oracle AI Vector Search, Private Agent Factory, and the Autonomous AI Database MCP Server
Where candidates lose points
This is an associate-level exam, but it spans several distinct tool ecosystems in one sitting - LangChain, MCP, the OpenAI Agents SDK, and OCI Enterprise AI Agents. Most lost points come not from any single module being hard, but from mixing up which framework owns which concept, like confusing LangChain's tool-calling flow with the OpenAI Agents SDK's handoff mechanism, or MCP's client/server roles with a generic REST integration. The Oracle AI Database module is also easy to underprepare for if your background is general AI engineering rather than Oracle-specific.
Sample question
A developer wants an AI agent to discover and call tools exposed by an independently maintained service without hardcoding that service's API contract into the agent's code. Which protocol is designed specifically for this kind of interoperable tool discovery?
- A. REST API versioning
- B. Model Context Protocol (MCP)
- C. OAuth 2.0 client credentials flow
- D. GraphQL federation
Full explanations and 119 more questions like this are in the practice tests.
How long to prepare
For someone who already writes Python and has touched an LLM API before, 2-3 weeks of focused study is a reasonable target. The 60-minute, 40-question format with a 65% pass mark rewards breadth over depth, so time is better spent skimming all six modules than mastering any single framework in isolation.
